The claim that humans share 50% or 60% of their DNA with bananas is misleading. It traces to a 2013 NHGRI analysis that found about 60% of human protein-coding genes have a recognizable counterpart in banana. But "60% of genes have a counterpart" is not "60% of DNA is shared."
Strict orthology methods put the real figure at 17-25% of human genes. Whole-genome alignment would be less than 1%, because most non-coding DNA has diverged beyond recognition over roughly 1.5 billion years.
How much DNA do humans share with bananas?
About 60% of human protein-coding genes have a recognizable counterpart in the banana genome. That figure comes from a 2013 analysis by NHGRI geneticist Dr. Lawrence Brody, prepared for a Smithsonian educational video - not a published paper. The video reported human and banana proteins as "41 percent similar."[1]
The 41% is the average amino-acid identity across roughly 7,000 matched protein pairs. Brody's team compared every predicted banana protein against every human protein using BLAST and averaged the hits. As he noted, "this is the average similarity between proteins (gene products), not genes."[1]
The popular "50%" soundbite conflates three different numbers: the share of human genes with a banana counterpart (~60%), the average amino-acid identity of those matches (~40%), and the share of total DNA that aligns (under 1%). A 2020 re-analysis by the Dessimoz Lab traced the claim through eight popular-science sources, most without any citation. Using three orthology methods - OMA, OrthoInspector, and best-bidirectional BLAST - the lab found that 17-25% of human genes have confirmed orthologs in banana, not 50%.[2]
The Swiss Institute of Bioinformatics reached a similar result with OMA: about 25% of human genes are shared with banana, across 3,400-4,900 orthologs.[3]
The banana genome itself has 36,542 protein-coding gene models across 523 megabases.[4]
How similar is human DNA to a banana?
The answer depends on what you measure. Three common methods give very different results for distantly related species.
- Whole-genome alignment lines up every DNA letter. By this measure, the human genome is 91% alignable with chimpanzee, 33% with mouse, and about 1% with zebrafish. For any plant, the fraction is less than 1% because non-coding DNA, which makes up over 98% of the human genome, accumulates mutations too fast to remain recognizable over hundreds of millions of years.[3][5]
- Ortholog counting measures protein-coding genes inherited from a common ancestor. This is the most meaningful comparison for distant species, because genes evolve more slowly than non-coding DNA. By this measure, humans share ~98% of genes with chimpanzee, ~94% with mouse, ~72% with zebrafish, and ~25% with banana.[3]
- Protein identity measures the average sequence similarity of the matched proteins. Brody's BLAST analysis found ~40% average amino-acid identity across the ~7,000 human-banana matches - the figure the Smithsonian video rounded to "41 percent similar."[1]
| Species | Shared genes (OMA orthologs) | Whole-genome alignment |
|---|---|---|
| Chimpanzee | ~98% | 91% |
| Mouse | ~94% | 33% |
| Zebrafish | ~72% | ~1% |
| Banana (plant) | ~25% | under 1% |
Gene-level comparisons use the OMA orthology method. Whole-genome alignment figures come from Ensembl Compara.[3][5]
How much DNA do humans share with fish, yeast, and bacteria?
There is no single percentage of DNA that humans share with fish, yeast, or bacteria. The clearest published comparisons count homologous genes or compare the proteins they encode, whereas a percentage of identical DNA letters would require a whole-genome alignment.
For fish, zebrafish (Danio rerio) provides a well-studied example. Its reference-genome paper found that 71.4% of human genes have at least one zebrafish ortholog. The 71.4% figure counts genes descended from genes in a common ancestor. Whole-genome identity is a separate measurement.[14]
For baker's yeast (Saccharomyces cerevisiae), more than one-third of yeast genes have human orthologs. Across those matched genes, the encoded proteins average 32% amino-acid identity, with individual pairs ranging from 9% to 92%. Both figures describe a selected set of genes or proteins, not whole-genome DNA identity.[15]
Bacteria are too diverse for a species-wide percentage, and even a named comparison such as human versus Escherichia coli depends on the search method. One study compared human proteins with 1,074 eubacterial proteomes and found 3,342 proteins with eubacterial homologs within the human protein-interaction network it analyzed. That result pools many bacterial species and uses a restricted set of human proteins, so it cannot be turned into a percentage of DNA shared with E. coli.[16]
The same distinction applies across all three comparisons: gene homology, protein identity, and whole-genome sequence identity answer different questions. A shared gene can remain recognizable even when most surrounding non-coding DNA no longer aligns.
Why do humans share so much DNA with bananas?
Humans and bananas share a common ancestor that lived roughly 1.576 billion years ago. A 1999 molecular-clock study by Wang, Kumar, and Hedges placed the three-way divergence of plants, animals, and fungi at 1,576 ± 88 million years ago. Every gene humans and bananas still share has survived that entire span.[6]
The surviving genes are overwhelmingly housekeeping genes - the ones needed for basic cellular functions that all eukaryotes share. A Gene Ontology analysis by the Dessimoz Lab found human-banana orthologs are enriched for cellular metabolism, gene expression, RNA processing, and lipid metabolism.[2] The Swiss Institute of Bioinformatics found the same pattern.[3]
This is why the shared percentage is not zero despite 1.5 billion years of separation. A cell is a cell, whether it sits in a human liver or a banana leaf. Both need to copy DNA, divide, produce energy, and build proteins. As Brody put it: "If you think about what we do for living and what a banana does there's a lot of things we do the same way, like consuming oxygen. A lot of those genes are just fundamental to life."[1]
How much DNA do humans share with other plants?
All plants diverged from the animal lineage at the same time, so the percentage of shared genes should be similar for any plant - banana, strawberry, potato, corn, or any other - in the range of 17-25%. The specific plant does not change the answer, because the evolutionary distance is the same.[6]
What varies is genome size and gene count. Plant genomes span three orders of magnitude, from the 30-megabase mushroom genome to the 16-gigabase onion genome, about five times larger than the human genome. But a larger genome does not mean more shared genes; the shared fraction depends on evolutionary distance, not genome size.
| Organism | Genome size | Protein-coding genes |
|---|---|---|
| Mushroom (Agaricus bisporus) | ~30 Mb | ~10,438 |
| Strawberry (Fragaria vesca) | ~240 Mb | ~34,809 |
| Banana (Musa acuminata) | ~523 Mb | ~36,542 |
| Cabbage (Brassica oleracea) | ~630 Mb | ~61,421 |
| Potato (Solanum tuberosum) | ~844 Mb | ~39,031 |
| Tomato (Solanum lycopersicum) | ~900 Mb | ~34,727 |
| Corn (Zea mays) | ~2,300 Mb | ~32,540 |
| Onion (Allium cepa) | ~16,000 Mb | ~50,000 |
Genome sizes and gene counts come from published genome sequencing studies.[4][7][8][9][10][11][12][13]
The internet is full of specific per-plant percentages - "humans share 70% with strawberries" or "50% with potatoes" - but these almost never come from primary sources. The Dessimoz Lab found that most popular-science articles reporting human-banana similarity lacked any citation at all.[2]
One exception is the mushroom. Mushrooms are fungi, not plants, and fungi are more closely related to animals than plants are. So humans share a higher percentage of genes with a mushroom than with a banana, despite the mushroom genome being much smaller.[12]
DNA-sharing comparisons depend on both DNA length and human gene count definitions, because shared genes and whole-genome alignment answer different questions.
Citation
ProteinIQ. "How much DNA do humans share with bananas?" Published June 30, 2026. Accessed [your access date]. https://proteiniq.io/guides/human-banana-dna-similarity
Sources▼
- Do People and Bananas Really Share 50 Percent of the Same DNA? HowStuffWorks · June 30, 2026. https://science.howstuffworks.com/life/genetic/people-bananas-share-dna.htm
- The Banana Conjecture Dessimoz Lab, University of Lausanne · 2020. https://lab.dessimoz.org/blog/2020/12/08/human-banana-orthologs
- Banana Split - Story Swiss Institute of Bioinformatics (In the Light of Evolution) · June 30, 2026. https://lightofevolution.org/en/banana-split/
- The banana (Musa acuminata) genome and the evolution of monocotyledonous plants Nature · 2012. https://www.nature.com/articles/nature11241
- Ensembl Compara: pairwise whole-genome alignments Ensembl · June 30, 2026. https://www.ensembl.org/info/genome/compara/analyses.html
- Divergence time estimates for the early history of animal phyla and the origin of plants, animals and fungi Proceedings of the Royal Society B · 1999. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1689654/
- The genome of woodland strawberry (Fragaria vesca) Nature Genetics · 2011. https://www.nature.com/articles/ng.740
- Genome sequence and analysis of the tuber crop potato Nature · 2011. https://www.nature.com/articles/nature10158
- The B73 maize genome: complexity, diversity, and dynamics Science · 2009. https://www.science.org/doi/10.1126/science.1178534
- The tomato genome sequence provides insights into fleshy fruit evolution Nature · 2012. https://www.nature.com/articles/nature11119
- The Brassica oleracea genome reveals the asymmetrical evolution of polyploid genomes Nature Genetics · 2014. https://www.nature.com/articles/ng.2965
- The genome of the button mushroom Agaricus bisporus Genome Biology · 2012. https://genomebiology.biomedcentral.com/articles/10.1186/gb-2012-13-5-r58
- A first assembly of the Allium cepa genome G3: Genes|Genomes|Genetics · 2021. https://academic.oup.com/g3journal/article/11/9/jkab243/6320787
- The zebrafish reference genome sequence and its relationship to the human genome Nature · 2013. https://www.nature.com/articles/nature12111
- Systematic humanization of yeast genes reveals conserved functions and genetic modularity Science · 2015. https://pmc.ncbi.nlm.nih.gov/articles/PMC4718922/
- The Human Genome Retains Relics of Its Prokaryotic Ancestry: Human Genes of Archaebacterial and Eubacterial Origin Exhibit Remarkable Differences Genome Biology and Evolution · 2011. https://pmc.ncbi.nlm.nih.gov/articles/PMC3163467/
Founder and computational chemist, ProteinIQ
Dr. Matic Broz is the founder of ProteinIQ and a computational chemist. He completed a PhD focused on protein structure, molecular dynamics, and neural networks, and writes about structural biology and scientific software.